NAD+ side effects: what trials and regulators have recorded

What side effects are reported with NAD+?

The side effects recorded for NAD+ depend on the form studied. A 2026 systematic review of 33 human intervention studies found nicotinamide riboside (NR) and nicotinamide mononucleotide (NMN) generally well tolerated over weeks to months, a review that calls for trials with longer follow-up and so cannot establish long-term safety (Gallagher and Emmanuel, Ageing Research Reviews, 2026). A 2026 meta-analysis whose safety analyses covered 10 adult randomised NMN trials found no rise in overall or serious adverse events against control groups, though the trials ran only 14 days to 24 weeks (Yang et al., Nutrients, 2026). In a crossover trial of 30 healthy middle-aged and older adults randomised, individual NR events were mostly mild, such as nausea, flushing and leg cramps, and several happened in placebo weeks, which limits what the trial can say about causes (Martens et al., Nature Communications, 2018). The US Food and Drug Administration (FDA) separately said it had received adverse event reports of severe chills, shaking, vomiting and fatigue following use of NAD+ drugs, reports that describe events without a rate (FDA notice, 30 October 2024).

The same word, NAD+, covers very different products, so one list of side effects would mislead. The table after these sections sets each form beside its record. The NAD+ research guide covers what the molecule is, and NAD+ is not a peptide settles the naming question.

Counts depend on who is counting. In a crossover trial of 30 healthy middle-aged and older adults randomised, 7 participants reported 14 events in all, every one mild: nine during the placebo weeks and five during the NR weeks, so the trial cannot say which events NR caused, and the events were self-reported in a small group (Martens et al., Nature Communications, 2018). In an 8-week supplier-funded trial of 140 overweight adults, 61 participants reported 95 events across a placebo group and three NR groups, with no serious events and no flushing, and a similar type, incidence and severity across groups, a short trial that cannot exclude rare or long-term harm (Conze et al., Scientific Reports, 2019).

Evidence limits

What have the safety studies not shown?

Trial design adds its own limits. Martens et al. gathered events by self-report at check-in visits and wrote that a cohort of 30 is too small to establish the broader safety profile of NR (Nature Communications, 2018). The 140-adult NR trial was funded by the supplier of the compound it tested, and two authors are employees of that supplier (Conze et al., Scientific Reports, 2019).

No source opened for this page reports how often any side effect happens in people who use NAD+ products outside trials. The FDA notice describes reports it received after use of NAD+ drugs and gives no denominator.

What do human studies conclude about the safety of NAD+ supplements?

A systematic review of 33 human intervention studies found NR and NMN generally well tolerated over weeks to months and judged clinical effectiveness for anti-aging or wellness outcomes inconclusive, and the review calls for trials with longer follow-up, so it cannot establish long-term safety (Gallagher and Emmanuel, Ageing Research Reviews, 2026). This review is not a safety finding for every person.

Trials enrolled specific groups. A 4-week NMN trial enrolled 31 healthy adults aged 20 to 65 (Fukamizu et al., Scientific Reports, 2022). A 12-week NR trial enrolled 40 sedentary obese men aged 40 to 70 (Dollerup et al., American Journal of Clinical Nutrition, 2018). A 30-day NR trial enrolled 30 newly diagnosed people with Parkinson's disease (Brakedal et al., Cell Metabolism, 2022). The trials described on this page enrolled adults.

The Therapeutic Goods Administration (TGA) requirements for listed medicines containing NMN specify an adult indication and no indication for pregnant or lactating women (TGA safety alert, last updated 3 February 2026); that is a regulatory requirement, not a safety finding. Is NAD+ legal in Australia? sets out that record.

What side effects did trials of nicotinamide mononucleotide (NMN) record?

In a 4-week trial of 31 healthy adults aged 20 to 65, five events were recorded (loose stool twice, a common cold, high blood pressure and acne), no severe adverse events were observed, and the investigators judged none of the five related to NMN, which a trial this small cannot confirm for rarer events (Fukamizu et al., Scientific Reports, 2022). A 12-week trial of 30 healthy adults (15 on NMN, 15 on placebo) found no abnormalities in physiological and laboratory tests and no obvious adverse effects, again in a small healthy group (Okabe et al., Frontiers in Nutrition, 2022).

In 80 healthy middle-aged adults followed for 60 days, with three NMN groups and a placebo group, no safety issues appeared in adverse events, laboratory or clinical measures, a result limited to that group and that length (Yi et al., GeroScience, 2023). In 30 overweight or obese adults aged 45 or older followed for 28 days, adverse events were similar between the NMN and placebo groups, in a small short trial (Pencina et al., Journal of Clinical Endocrinology and Metabolism, 2023).

Each of these trials is small and short. The meta-analysis in the next sections covers 10 adult randomised trials with 383 participants in its safety analyses (Yang et al., Nutrients, 2026).

Is NMN toxic to the liver?

A 2026 meta-analysis of adult randomised NMN trials found that NMN did not significantly raise ALT or AST, the two blood enzymes used to check for liver stress, in analyses of 10 comparisons with 330 participants over 14 days to 24 weeks, which cannot establish long-term liver safety (Yang et al., Nutrients, 2026). In a 4-week trial of 31 healthy adults, biochemical measures did not change beyond physiological variation, in a small healthy group that cannot exclude rare or long-term harm (Fukamizu et al., Scientific Reports, 2022). In Sprague-Dawley rats given NMN for 90 days, organ toxicity checks and tissue examinations showed no toxic effects, a rat result that does not establish human safety (Cros et al., Food and Chemical Toxicology, 2021).

A second rat study gave NMN to Sprague-Dawley rats for 91 days and found no test-item-related adverse findings at its lowest amount, while body weight, weight gain and food intake fell in the highest group, a pattern also seen with the nicotinamide riboside reference; the result is from rats and not from people (Turner et al., Frontiers in Pharmacology, 2021).

The related compound nicotinamide behaves differently. Memorial Sloan Kettering's About Herbs page lists liver toxicity among the effects associated with high amounts of nicotinamide, and says niacin and nicotinamide are not interchangeable (page last updated 21 February 2023). NMN is a different molecule, and the trials above are the evidence for NMN itself.

What side effects did trials of nicotinamide riboside record?

In an 8-week randomised trial of 140 overweight adults, 95 adverse events were reported by 61 participants, none serious, with no flushing and a similar type, incidence and severity across the placebo and three NR groups; the trial was supplier-funded and short (Conze et al., Scientific Reports, 2019). In a 12-week randomised trial of 40 obese men, no serious adverse events were attributed to NR and safety blood tests were normal, in a single group of men (Dollerup et al., American Journal of Clinical Nutrition, 2018).

In the NR crossover trial of 30 healthy middle-aged and older adults randomised, the events listed under NR were one report each of nausea, skin rash, flushing, leg cramps and increased bruising, in a small group over two 6-week periods (Martens et al., Nature Communications, 2018). In a 30-day phase 1 trial of 30 newly diagnosed people with Parkinson's disease, NR was well tolerated, a small short trial in one condition (Brakedal et al., Cell Metabolism, 2022). In a 4-week phase 1 trial of 20 people with Parkinson's disease (10 on high-amount NR, 10 on placebo), no moderate or severe adverse events were recorded, mild events did not differ significantly between groups, and serum homocysteine rose slightly early on in NR recipients, again in a small group (Berven et al., Nature Communications, 2023).

Flushing was tracked in two NR trials. In an 8-week supplier-funded trial of 140 overweight adults, Conze et al. reported no flushing with NR, a short trial that cannot exclude rare events (Scientific Reports, 2019). In a crossover trial of 30 healthy middle-aged and older adults, Martens et al. recorded one flushing event in the NR weeks against two in the placebo weeks, a small trial with self-reported events (Nature Communications, 2018).

What adverse events have been recorded after NAD+ is given by IV?

In a pilot of 11 men, eight given NAD+ and three given saline, no adverse events were observed, and a group this small cannot show how often events occur (Grant et al., Frontiers in Aging Neuroscience, 2019). The FDA says it received adverse event reports of severe chills, shaking, vomiting and fatigue following use of NAD+ drugs, some requiring medical treatment, reports that carry no rate (FDA notice, 30 October 2024).

In a warning letter dated 20 January 2026, the FDA describes a complaint about three patients who developed low blood pressure, uncontrollable shaking, shivers and body aches shortly after or during a compounded NAD+ product, and who were directed to an emergency room; this is an inspection record about one facility and not a controlled study (FDA warning letter 718739, 20 January 2026). The FDA notice calls the reactions it received consistent with excessive levels of endotoxins, which are fragments of bacterial cell walls (FDA notice, content current as of 30 October 2024).

The two records describe different things: a small pilot with a saline comparison group, and reports the FDA gathered after clinical use of compounded products. A systematic review also found one nonrandomised NMN study, which mainly added short-term safety and biomarker information (Gallagher and Emmanuel, Ageing Research Reviews, 2026). No source opened here reports how common any side effect is for compounded NAD+ products.

What has the FDA said about compounded NAD+ products?

The FDA has warned compounders about NAD+ in two documents read for this page. A notice with content current as of 30 October 2024 says the agency is aware of compounders using food-grade NAD+ sold by repackagers, and that ingredients identified as food grade are not suitable for compounding sterile drugs without appropriate processing, because of the high risk of contamination with microbes and endotoxins. A warning letter dated 20 January 2026 to one US outsourcing facility states that NAD+ does not appear on the 503B bulks list and that an unopened vial from one lot of its NAD+ product contained excessive bacterial endotoxins; the letter concerns that facility and that lot (FDA warning letter 718739).

The letter records the finding in detail. An unopened vial from the lot tested at 3,360 endotoxin units per millilitre, and the firm began a voluntary recall of its sterile drug products on 30 July 2025 (FDA warning letter 718739, 20 January 2026). Compounders are pharmacies and outsourcing facilities that prepare custom drugs, and the 503B bulks list names the bulk drug substances with a clinical need that an outsourcing facility may use in compounding (FDA warning letter 718739).

The letter concerns one outsourcing facility and its lots. It is not a statement about every NAD+ product. NAD+ legal status in Australia covers the Australian side.

How long does NAD+ stay in the blood after an IV infusion?

The only human data on plasma NAD+ after NAD+ was given that were opened for this page come from a 2019 pilot, and the paper reports no half-life. In that pilot of eight men given NAD+, plasma NAD+ and its breakdown products did not change until after the 2-hour mark, and the authors conclude that NAD+ was rapidly and completely removed from plasma for at least the first 2 hours under the conditions tested; the paper gives no clearance time (Grant et al., Frontiers in Aging Neuroscience, 2019). In the eight men given NAD+ in Grant et al.'s 2019 pilot, six-hour urine samples showed increased methylnicotinamide and NAD+; this small metabolite study did not establish a whole-body clearance time.

The limits are large. The test group was eight men aged 30 to 55 with a body mass index under 30, studied under one set of conditions, with three saline controls (Grant et al., Frontiers in Aging Neuroscience, 2019). The paper shows what appeared in plasma and urine in that one pilot. It does not give a clearance time, and it does not cover other forms of NAD+.

What risks have researchers raised about raising NAD+ levels?

Researchers have listed potential risks of raising NAD+ levels, and the human trials above lasted only weeks to months. A systematic review of 147 articles, 113 preclinical and 34 clinical, described a limited acute toxicity profile for NAD+ precursors and named potential risks as accumulation of putative toxic metabolites, tumorigenesis and promotion of cellular senescence, adding that evidence is limited (Braidy and Liu, Experimental Gerontology, 2020). The review presents these as potential risks, and they are not observed outcomes in the human trials above.

A separate line of work concerns niacin, not NAD+ itself. In two validation cohorts of stable cardiac patients (2,331 and 832 people), blood levels of 2PY and 4PY, the terminal metabolites of excess niacin, were associated with higher 3-year risk of major adverse cardiovascular events, and 4PY induced a vascular adhesion protein in mice, a mouse result that does not establish a human result; the study measured niacin metabolites and not NMN or NR, so it does not show that an NAD+ precursor raises that risk (Ferrell et al., Nature Medicine, 2024).

The TGA alert states that products making claims about NAD, NAD+, NADH or NMN that do not have an AUST number on the label may be unapproved goods or illegally supplied and could pose a risk to health (TGA safety alert, last updated 3 February 2026).

What do official medical sources say about NAD+ side effects?

The official sources opened for this page cover different products, so their records cannot be treated as one safety finding for NAD+. The TGA says products making claims about NAD, NAD+, NADH or NMN that do not have an AUST number on the label may be unapproved goods or illegally supplied and could pose a risk to health (TGA safety alert, last updated 3 February 2026). The FDA notice covers NAD+ ingredients used by compounders (FDA notice, content current as of 30 October 2024). Memorial Sloan Kettering's About Herbs page lists nausea, vomiting, headache, fatigue, dizziness and liver toxicity as associated with high amounts of nicotinamide, a related vitamin; it describes nicotinamide and not NAD+ (page last updated 21 February 2023).

A search of the Memorial Sloan Kettering site on 14 October 2026 returned no separate entry for nicotinamide riboside. The trial reports and reviews cited on this page are indexed in PubMed, the US National Library of Medicine database, and they are the main published evidence for NR and NMN found in the sources searched.

An ARTG number, which appears on the label as AUST L, AUST L(A) or AUST R, shows that a medicine is on the Australian Register of Therapeutic Goods (TGA safety alert, last updated 3 February 2026).

Is research-use NAD+ the same material these studies measured?

No. Every safety figure on this page was measured on a trial product or recorded in clinical use, and none describes research-use reference material. Research material is not a medicine, is not a substitute for one, and is not for human or animal use. The trial products were identified supplies given under study protocols, and one NR trial was funded by the supplier of the compound it tested (Conze et al., Scientific Reports, 2019). The trial results describe what happened inside those trials and say nothing about any material outside them.

The site catalogue lists NAD+ reference material as a laboratory item. What a batch document can and cannot show is a separate subject from what a trial measured, and the two should not be merged.